daily-paper-generator

Searches arXiv for recent EEG decoding papers and generates bilingual structured paper summaries.

Updated Oct 7, 2022
One-click install
npx skills add https://github.com/tamagusko/linux-cfg --skill daily-paper-generator-tamagusko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: daily-paper-generator
Source: https://github.com/tamagusko/linux-cfg/tree/main/dotfiles/claude/skills/daily-paper-generator
Command: npx skills add https://github.com/tamagusko/linux-cfg --skill daily-paper-generator-tamagusko

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires feedparser, and includes scripts (resource) and references (resource) components.

What problem does it solve? Manually tracking new arXiv publications on EEG decoding, speech decoding, and brain foundation models is time-consuming, and writing consistent bilingual paper reviews requires significant effort. This Skill automates the discovery, quality evaluation, and summarization of recent papers into a standardized Chinese/English Markdown format. ## Core Features & Use Cases - arXiv Paper Discovery: Searches arXiv via Chrome browser automation using curated keyword lists for EEG decoding, EEG speech decoding, and brain foundation models, filtered to the last 3 months. - Structured Quality Evaluation: Scores papers on a 5-dimension rubric (innovation, method completeness, experimental thoroughness, writing quality, relevance) and selects the top 3. - Bilingual Summary Generation: Produces complete Markdown summaries with a ~300-word Chinese review, an academic English review, metadata tables, and an integrated publishing format. - Use Case: A neuro-AI researcher asks to "generate daily paper" each morning and receives three timestamped Markdown files in a daily paper/ directory, each containing a fully formatted bilingual review ready for publication. ## Quick Start Ask the assistant to generate today's daily paper by searching arXiv for recent EEG decoding papers and writing bilingual summaries of the top three.

Frequently Asked Questions about daily-paper-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search arXiv for recent EEG decoding papers?

Use the arXiv search URL with keywords like "EEG decoding" or "EEG foundation model", sorted by announced date descending. This Skill automates that search through Chrome browser automation and filters results to papers from the last 3 months.

How are papers evaluated and selected for the daily summary?

Each candidate paper is scored on five weighted dimensions: innovation (30%), method completeness (25%), experimental thoroughness (25%), writing quality (10%), and relevance (10%). Papers are ranked by weighted total score and the top 3 are selected for summarization.

What output format does the daily paper summary use?

Each summary is a Markdown file with nine sections: title, authors and institution, arXiv link with metadata, a ~300-word Chinese review, an English review, a main figure placeholder, a metadata table, an integrated publishing format, and an appendix with sources.

Does this workflow require Chrome browser access?

Yes, the primary workflow uses Chrome MCP tools to navigate arXiv search and paper pages and extract metadata. A legacy Python script using the arXiv API and feedparser exists in scripts/ but is marked deprecated.

Can I search topics other than EEG decoding?

The keyword lists in references/keywords.md cover EEG decoding, EEG speech decoding, and brain foundation models, but the same search and review workflow can be applied to other arXiv topics by supplying different query keywords.